Large Number Estimation as a Vehicle to Promote Mathematical Modeling

This study presents a teaching experiment in which second-grade primary school students compared a city and a town according to population estimates. The activities were presented to them as Fermi problems and required an analysis of the reality to identify sub-problems that students could deal with. The students' products were analyzed from the perspective of mathematical modeling and it was observed that the students developed their own mathematical methods to make the required estimates. Learning opportunities were identified when students needed concepts and procedures that they had not yet worked with. The results show that estimation acts as a facilitator for the generation of mathematical models that enable students to connect their real-world knowledge to new mathematical learning. In the conclusions, there is a discussion of the implications of the use of Fermi problem sequences to initiate mathematical modeling and link up mathematical and extra-mathematical knowledge about social problems in the first years of primary education.